Article pubs.acs.org/molecularpharmaceutics

Mechanism Study of Cellular Uptake and Tight Junction Opening Mediated by Goblet Cell-Specific Trimethyl Chitosan Nanoparticles Jian Zhang, Xi Zhu, Yun Jin, Wei Shan, and Yuan Huang* Key Laboratory of Drug Targeting and Drug Delivery System, Ministry of Education, West China School of Pharmacy, Sichuan University, No. 17, Block 3, Southern Renmin Road, Chengdu 610041, P. R. China ABSTRACT: Modifying nanoparticles with targeting peptides which can specifically bind to intestinal epithelium was recently suggested as a strategy to further enhance their ability for the oral delivery of macromolecular drugs. However, few studies were focused on comprehensive understanding of the uptake and transport processes as well as the underlying molecular signaling pathways mediated by the ligand modification. In the present study, the mechanisms of cellular uptake and the tight junction opening associated with the trimethyl chitosan based nanoparticles (M-NPs) and their goblet cell-targeting CSK (CSKSSDYQC) peptide modified nanoparticles (CSK-M-NPs) were investigated. Compared with single ion crosslinked nanoparticles (S NPs), M-NPs and CSK-M-NPs, prepared with multiple agents, exhibited superior stability which could effectively protect drugs against the degradation of trypsin. Caveolae-mediated endocytosis and macropinocytosis were involved in the intracellular uptake of both M-NPs and CSK-M-NPs on Caco-2/HT29-MTX cocultured cells. However, CSK peptide modification could further induce clathrin-mediated endocytosis of the NPs. Intriguingly, most endocytosis subpathways have been altered after CSK peptide modification. Moreover, the opening of epithelial tight junctions was investigated at both protein and gene levels. The results indicated that both M-NPs and CSK-MNPs could transiently and reversibly open the epithelial tight junctions via the C-Jun NH2-terminal kinase-dependent pathway. However, CSK peptide modification enabled a more rapid opening and recovering of the tight junctions. In all, the enhanced uptake and transport capacity of nanoparticles after CSK peptide modification may be attributed to the alteration of internalization pathways and the stronger ability of opening tight junctions. KEYWORDS: goblet cell-specific, trimethyl chitosan nanoparticles, cellular uptake, tight junctions, mechanism

1. INTRODUCTION Oral delivery of macromolecular drugs remains a challenge due to their poor enzymatic stability in the gastrointestinal (GI) tract and low permeability across the intestinal epithelium. Chitosan (CS) is one of the most widely explored natural polymers for oral delivery of macromolecules. It has some favorable properties such as high biocompatibility, low cost, and convenience for modification.1 Chitosan based nanoparticles (CS NPs) can protect the loaded biotherapeutics against the harsh environment in the GI tract and simultaneously prolong the residence time by mucoadhesion.2 In addition, CS NPs can mediate the opening of tight junctions (TJs) between epithelial cells and, thus, facilitate the paracellular permeation of hydrophilic macromolecules.3 Intestinal epithelium is the primary physiological absorption barrier, which regulates the uptake of the orally administered therapeutic agents. The transportation of substrates (transcellular transport or paracellular transport) through epithelium is a complicated process, which involves multiple pathways with different mechanisms. The uptake of nanoparticles (NPs) by epithelial cells could be influenced by many factors, including the particle size,4 surface charge,5 hydrophobicity,6 and even concentration of nanoparticles.7 Incorporation of targeting © 2014 American Chemical Society

moieties on the particle surface could enhance their affinity to some specific sites of the epithelium8 and influence the mode of drug absorption. Ligands, such as folic acid and albumin, have been shown to facilitate the uptake of particles through caveolin-mediated endocytosis, whereas ligands for glycol receptors promote clathrin-mediated endocytosis.9 Our group has recently developed insulin-loaded single ion cross-linked trimethyl chitosan nanoparticles modified with goblet celltargeting CSK peptide. The modified nanoparticles were confirmed to enhance the intestinal absorption of insulin by targeting to the epithelium goblet cells.10 However, the internalization and the subsequent transfer pathways of these NPs associated with enterocytes and goblet cells have not been demonstrated. In addition, the permeation of drugs from the intestinal lumen to the systemic circulation could also be achieved by the transient opening of TJs mediated by chitosan.3 This process involves the redistribution of actin filaments, transmembrane integral proteins (such as claudins and Received: Revised: Accepted: Published: 1520

November 13, 2013 March 19, 2014 March 27, 2014 March 27, 2014 dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Table 1. Size, Zeta Potential, Encapsulation Efficiency (EE %), and Drug Loading (DL %) of NPs Prepared under Varying Concentrations of TMC, TPP, MgSO4, and γ-PGA-Phe (Mean ± SD, n = 3) NPs

insulin (mg/mL)

TPP (mg/mL)

MgSO4 (mg/mL)

γ-PGA-Phe (mg/mL)

TMC (mg/mL)

1 2 3 4 5 6 7 8 9 10 11

0.136 0.152 0.168 0.152 0.152 0.152 0.152 0.152 0.152 0.152 0.152

0.32 0.32 0.32 0.28 0.36 0.32 0.32 0.32 0.32 0.32 0.32

0.26 0.26 0.26 0.26 0.26 0.24 0.28 0.26 0.26 0.26 0.26

0.105 0.105 0.105 0.105 0.105 0.105 0.105 0.1 0.11 0.105 0.105

0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.56 0.72

size (nm) 240.7 212.5 300.2 290.3 270.7 303 320 290.3 240.2 267.3 252.4

± ± ± ± ± ± ± ± ± ± ±

2.3 3.6 3.5 1.2 0.3 3.1 2.6 4.4 5.4 3.8 4.1

zeta potential (mV) 23.35 22.31 20.21 23.24 18.62 18.45 24.14 21.09 22.12 20.25 28.25

± ± ± ± ± ± ± ± ± ± ±

0.68 0.68 2.31 1.35 1.03 1.57 1.35 2.05 0.72 1.24 0.03

EE % 44.46 52.78 50.15 39.12 52.24 45.3 56.13 51.24 42.91 40.25 41.38

± ± ± ± ± ± ± ± ± ± ±

3.25 3.21 0.97 3.05 2.01 2.07 1.67 3.01 3.17 1.82 2.19

DL % 27.31 35.26 24.32 36.12 30.19 32.14 40.26 28.68 27.29 32.39 40.26

± ± ± ± ± ± ± ± ± ± ±

2.43 1.27 2.64 1.78 0.16 0.26 1.06 1.81 0.74 4.2 3.21

Sigma-Aldrich (St. Louis, MO, USA). Alexa Fluor 594-labeled dextran conjugate (10,000 MW), Alexa Fluor 594-labeled cholera toxin beta subunit (CTB), transferrin from human serum, Alexa Fluor 594 conjugate, and mouse monoclonal anticlaudin-4 were purchased from Molecular Probes (Eugene, OR, USA). Rhodamine-conjugated goat anti-mouse IgG and normal goat serum were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). All other chemicals used were analytical grade. 2.2. Synthesis of Polymers. Trimethyl chitosan (TMC) was synthesized by methylation of amine groups of chitosan with methyl iodide in N-methylpyrrolidone. The reaction was proceeded for 45 min at 60 °C. The product was purified by dialysis and then lyophilized (SNL216 V, Savant Instruments Inc., NY, USA). The obtained TMC polymers were dissolved in D2O for 1H nuclear magnetic resonance analysis (1H NMR, UNITY INOVA-400, Varian Inc., Palo Alto, CA, USA). The degree of quaternization was calculated from the integration of 1 H NMR according to the previously described method using the equation14

occludin), and junctional adhesion molecules. Although these mechanisms have been discussed by some researchers,11−13 no investigation regarding ligand modified nanoparticles with targeting property has been reported. Therefore, investigation on the interactions between ligand modified CS NPs and intestinal epithelial cells is very important for the active oral delivery of macromolecular drugs. In the present study, the internalization and transportation mechanisms of CSK peptide modified trimethyl chitosan nanoparticles were investigated. Since poor colloidal stability of single ion cross-linked nanoparticles (S NPs) severely limited their application, novel insulin loaded trimethyl chitosan nanoparticles (M-NPs) with improved stability were prepared for further investigations. For the preparation of M-NPs, cationic trimethyl chitosan (TMC) was cross-linked with sodium tripolyphosphate (TPP), magnesium sulfate (MgSO4), and glutamic acid grafted L-phenylalanine ethyl ester (γ-PGA-Phe). Goblet cell-targeting CSK peptide was conjugated to the TMC materials for the preparation of CSK peptide modified NPs (CSK-M-NPs). The drug stability against enzyme, cellular uptake mechanisms, and the underlying molecular signaling pathways regarding the effects of NPs on the opening of TJs were explored. Caco-2/HT29-MTX cocultured cells, which consist of both absorptive enterocytelike Caco-2 cells and the mucus-producing goblet cell-like HT29-MTX cells, were used. The contribution of this article is the systematic study of the influence of ligand modification on the uptake and transport processes of TMC-based NPs across epithelial cells, including endocytosis, intracellular trafficking, transcytosis, and the effect on TJs.

% DQ = {[(CH3)3 ]/[H] × 1/9} × 100

where % DQ is the degree of quaternization in percentage, [(CH3)3] is the integral of the chemical shift of the trimethyl amino group at 3.3 ppm, and [H] is the integral of the 1H peaks between 4.7 and 5.7 ppm. CSK peptide conjugated TMC (TMC−CSK) was synthesized by the formation of an amide bond between the residual primary amino groups on TMC and carboxyl groups on CSK peptide as previously described.10 Briefly, TMC (0.27 mM of primary amino groups), 1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride (EDC·HCl) (0.45 mM), and Nhydroxysuccinimide (NHS) (0.45 mM) were dissolved in 12 mL of distilled water and the vessel was filled with nitrogen. Then the CSK peptide was added with a final concentration of 0.045 mM. The reaction was conducted at room temperature for 3 days in the dark. Subsequently, the product was dialyzed, lyophilized, and stored at 4 °C. The obtained TMC−CSK was identified by 1H NMR and Fourier transform infrared (FT-IR) spectra (VECTOR22, Bruker, Germany). The content of the conjugated peptide was determined through amino acid detection (835-50, Hitachi Co., Tokyo, Japan). γ-Poly glutamic acid (γ-PGA) grafted with L-phenylalanine ethyl ester ( L-Phe-OEt) was synthesized as previously described.15 Briefly, γ-PGA (4.0 mM) was hydrophobically modified with L-Phe-OEt (4.0 mM) in the presence of EDC·

2. MATERIALS AND METHODS 2.1. Materials. Chitosan (deacetylation degree >90% and molecular weight of 400 kDa) was provided by AK Biotech Co., Ltd. (Shandong, China). CSKSSDYQC (CSK) peptide was chemically synthesized by Chinese Peptide Co., Ltd. (Hangzhou, China). 1-[3-(Dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride (EDC·HCl) was gained from Meapeo Co., Ltd. (Shanghai, China). N-Hydroxysuccinimide (NHS), N-methylpyrrolidone, iodomethane, and acetonitrile were all obtained from Kelong Chemical Co., Ltd. (Chengdu, China). γPoly glutamic acid was gained from PTC Co., Ltd. (Guangzhou, China). Porcine insulin was purchased from Wanbang Bio-Chemical Co., Ltd. (Jiangsu, China). Fluorescein isothiocyanate (FITC), genistein, dynasore, cytochalasin D, wortmannin, and other inhibitors were all purchased from 1521

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

HCl (4.0 mM) in sodium hydrogen carbonate aqueous solution (50 mM). Then the product was purified by dialysis and was subsequently lyophilized. The obtained γ-PGA-Phe was characterized by 1H NMR. The grafting degree of L-Phe-OEt was determined from the integral intensity ratio of the methylene peaks of γ-PGA to the phenyl group peaks of LPhe-OEt. Moreover, FITC-labeled insulin (FITC-insulin) was prepared based on the reaction between the isothiocyanate groups of FITC and the primary amino groups of insulin as reported.16 2.3. Preparation of NPs. 2.3.1. Preparation of S NPs. The insulin loaded single ion cross-linked NPs (S NPs) were prepared by self-assembly of TMC and TPP.1 Briefly, insulin was dissolved in HCl (pH 2.0) at a concentration of 1 mg/mL, and the pH was adjusted to 8.0 using NaOH solution (1 M). The insulin solution was premixed with TPP solution (1 mg/ mL) at a ratio of 1:2.5 (v/v). The premixed solution was then added dropwise to TMC aqueous solution (1 mg/mL) at an equal volume. The mixture was stirred at room temperature for 20 min, yielding an opalescent suspension. The resultant suspension was centrifuged at 14,000 rpm at 4 °C (Allegra X22R, Beckman Coulter Inc., Pasadena, CA, USA). The supernatants were discarded and the nanoparticles were resuspended for further use. 2.3.2. Preparation of M-NPs or CSK-M-NPs. Insulin loaded trimethyl chitosan nanoparticles (M-NPs) as well as the ligand modified NPs (CSK-M-NPs) were prepared with TMC or TMC−CSK and multiple cross-linking agents including γ-PGAPhe, TPP, and MgSO4.17 The NP formulation was optimized according to size, zeta potential, encapsulation efficiency (EE), and drug loading (DL). The influence of the amounts of TMC, insulin, TPP, MgSO4, and γ-PGA-Phe on the characteristics of nanoparticles was investigated (Table 1). In brief, equal volumes of the insulin solution were premixed with aqueous solutions of γ-PGA-Phe. Subsequently, TPP and MgSO4 were blended into the mixture, which was kept stirring for 1 h. The mixed solutions were then added dropwise to TMC or TMC− CSK aqueous solution under magnetic stirring at room temperature, yielding an opalescent suspension. The suspension was ultracentrifuged at 14,000 rpm (30 min, 4 °C). The supernatant was discarded, and the obtained nanoparticles were resuspended for further use. The optimized formulation was used for the preparation of the M-NPs or CSK-M-NPs. Fluorescence-labeled NPs were prepared with FITC-insulin following the same procedure. 2.4. Characterization of NPs. The size and zeta potential of prepared NPs were measured by a Malvern Zetasize NanoZS90 (Malvern Instruments Ltd., Malvern, U.K.). For the evaluation of the EE and DL, the NP suspension was ultracentrifuged at 14,000 rpm at 4 °C for 30 min. For the insulin-loaded NPs, the free drug in supernatant was measured by a reverse-phase high performance liquid chromatography (RP-HPLC) method (Agilent 1200 series, Santa Clara, CA, USA).18 For the FITC-insulin loaded batch, the free drug in supernatant was measured by fluorospectrophotometer (Shimadzu Corp., Tokyo, Japan). The excitation and emission wavelengths were set at 488 and 516 nm, respectively. The EE % and DL % of NPs were calculated as follows:

DL % = (total amount of insulin − free insulin) /(nanoparticle weight) × 100

2.5. Drug Release Study. For the evaluation of drug release of different NP formulations, 7 batches of NPs were prepared according to the optimized formulation, some of which were prepared with the absence of one component (Table 2). Drug release study was performed in phosphate Table 2. NP Formulation with Different Combinations of Agents components NPs S NPs M-NPs NPs1 NPs2 CSK-M-NPs NP3 NP4

insulin insulin insulin insulin insulin insulin insulin

TPP TPP TPP TPP TPP TPP TPP

MgSO4 MgSO4 MgSO4 MgSO4

γ-PGA-Phe γ-PGA-Phe γ-PGA-Phe γ-PGA-Phe

TMC TMC TMC TMC TMC−CSK TMC−CSK TMC−CSK

buffered saline (PBS) at pH = 6.8, and all samples were kept at 37 °C and shaken at 100 rpm. 100 μL samples were withdrawn at predetermined time intervals (0, 0.5, 1, 2, 4, 8, 12 h), and then equal volumes of PBS were supplemented. Subsequently, the withdrawn samples were centrifuged (14,000 rpm, 30 min, 4 °C) and the supernatant was analyzed using the RP-HPLC method. 2.6. Proteolytic Stability Test. All tested NPs and the free insulin solution were suspended in simulated intestinal media with trypsin (0.2%, w/v) and incubated at 37 °C. Enzymatic degradation of the samples was immediately stopped at predetermined time intervals (0, 0.5, 1, 2, 3, 4, 6, and 8 h) by adding 0.1 M HCl solution. The amount of remaining insulin was determined by RP-HPLC analysis. The percentage of change from the initial amount of insulin was plotted against time. Tests were performed in triplicate for each sample. 2.7. Cell Studies. 2.7.1. Cell Culture. Caco-2 cells were gained from Institute of Biochemistry and Cell Biology (Shanghai, China). HT29-MTX cells were a kind gift from Dr. Thecla Lesuffleur (INSERM, Paris, France). They were both cultivated separately in 75 cm2 culture flasks using Dulbecco’s modified Eagle medium (DMEM, Gibco, NY, USA) containing 10% fetal calf serum, 1% nonessential amino acid, 1% penicillin, and streptomycin (100 IU/mL) (Hyclone, Logan, UT, USA). Both cultures were maintained at 37 °C, 95% relative humidity and 5% CO2.19 Prior to the tests, cells were harvested using trypsin (0.25%) containing ethylenediamine tetraacetic acid (EDTA, 0.05 mM), respectively. Then, the Caco-2/HT29-MTX cocultured cells (7:3) were seeded onto 96-well plates (Corning, NY, USA) at a density of 5 × 104 cells/mL for cytotoxicity and uptake mechanism assays. The cocultured cells were incubated for 7 days before internalization studies. For transport assays, the cocultured cells were seeded onto the Transwell chambers consisting of polycarbonate membrane (0.4 μm in pore size, 0.33 cm2 of cell growth area, Costar) at a density of 3 × 104 cells/well, respectively. Cells were cultured for 21 days before use. 2.7.2. Cytotoxicity Assay. The cytotoxicity of the tested NPs or inhibitors was evaluated using 3-(4,5-dimethylthiazol-2-y1)-

EE (%) = (total amount of insulin − free insulin) /(total amount of insulin) × 100 1522

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

2,5-diphenyltetrazolium bromide (MTT) assay on Caco-2/ HT29-MTX cocultured cells. Prior to the test, the medium in 96-well plates was removed. Subsequently, cells were washed with Hanks balanced salt solution (HBSS) and incubated with tested NP solutions (at an insulin dose from 50 to 400 μg/mL) or inhibitor solutions (at adopted concentrations) in HBSS and then submitted to MTT assay. The group incubated with HBSS was used as control for 100% cell viability.20 2.7.3. Cellular Uptake Studies. To quantify the cellular uptake of NPs, cocultured cells were incubated with the fluorescence-labeled NPs (at a concentration of 300 μg/mL) for 3 h, and then washed twice with PBS. Cell lysis was achieved with lysis buffer containing 1% Triton X-100, 50 mM HEPES (pH 7.5), 150 mM NaCl, 2 mM Na3VO4, 100 mM NaF, 100 U/mL aprotinin, 20 μM leupeptin, and 0.2 mg/mL phenylmethanesulfonyl fluoride. Then the cell associated fluorescence and protein were determined by Varioskan Flash (Thermo Fisher Scientific, MA, USA) and bicinchoninic acid (BCA) assay kit (KeyGen Biotech Co., Ltd., Nanjing, China), respectively. The amounts of cellular uptake were expressed as the quantity of FITC-insulin associated with 1 mg of cellular protein. 2.7.4. Fluorescence Microscopy. Microscopy studies were used to investigate the internalization mechanism of both NPs. The cocultured cells were seeded onto 6-well plates (Corning, Corning, NY, USA) at a density of 2 × 104 cells/mL. The cell cocultures were incubated with fluorescence-labeled NPs and different endocytic markers: Alexa Fluor 594-labeled dextran (0.5 mg/mL), transferrin (50 μg/mL), and cholera toxin beta subunit (2 μg/mL, CTB), respectively.21−23 Then the cocultured cells were observed by inverted fluorescence microscope (XD30-RFL, Zhejiang, China). 2.7.5. Endocytosis Pathways. To further identify the internalization pathways of the fluorescence-labeled NPs through Caco-2/HT29-MTX cocultured cells, the cells were first cultured with different specific endocytic inhibitors (Table 3) for 30 min at 37 °C.24 After that, cellular uptake studies were performed in the presence of the same concentration of agents and NPs (at a FITC-insulin dose of 300 μg/mL) for 3 h at 37 °C. Thereafter, the test solutions were removed, and the cells were washed thrice with PBS. The amount of FITC-insulin in cocultured cells was tested as mentioned above. Their counterparts in the absence of inhibitors were used as control. 2.7.6. Transport Studies across Caco-2/HT29-MTX Cocultured Cell Monolayer. Transport studies were performed on Caco-2/HT29-MTX cocultured cell monolayer. The integrity of the monolayer was checked by measuring the transepithelial electrical resistance (TEER) values using a Millicell-ER system (Millipore Corporation, Bedford, MA) in HBSS buffer before the experiment. Only cells with TEER values in the range of 400−600 Ω/cm2 were used for transport studies.25 Briefly, filter inserts were rinsed with HBSS and allowed to equilibrate at 37 °C for 30 min. Experiments were carried out by replacing the apical buffer with 100 μL of fluorescence-labeled NPs (at a FITC-insulin dose of 300 μg/mL) in transport buffer at 37 °C for 3 h. Then 200 μL samples were collected from the basolateral chamber at predetermined times (0.5, 1, 1.5, 2, and 3 h) and equal volumes of fresh buffer were added. The amount of transported FITC-insulin from tested NPs was determined by fluorospectrophotometer, and the accumulative transport of FITC-insulin was calculated. Each experiment was performed in triplicate.

Table 3. Summary of the Results Obtained by Quantification of NPs Internalization in Caco-2/HT29-MTX Cocultured Cells upon Interfering with Known Pathways inhibition (%)

cellular environment

concn

heparinases

5U

amiloride

dynasore

12 μg/ mL 10 μM 5 μg/mL 0.06 mM 10 μg/ mL 0.05 mM

filipin

500 nM

cholesterol oxidase

100 mU

genistein

0.2 mM

chlorpromazine

0.3 mg/mL 0.15 mM

rottlerin cytochalasin D wortmannin lovastatin

monodansylcadaverine CSK peptide M-β-CD

0.02 mg/mL 1 μg/mL

MNPs

CSKMNPs

adsortive mechanisms macropinocytosis

70*a

80*

25*

24*

macropinocytosis macropinocytosis macropinocytosis caveolae-mediated endocytosis caveolae-/clathrinmedicated endocytosis and lipid rafts caveolae-mediated endocytosis caveolae-mediated endocytosis caveolae-mediated endocytosis clathrin-mediated endocytosis clathrin-mediated endocytosis receptor mediated endocytosis lipid rafts and lipidraft-mediated endocytosis

44* −b 77* 55*

57* 54* − 38*

55*

59*

26*

−

−

70*

29*

64*

−

69*

−

60*

−

50*

30*

−

associated pathway

a (*) Significant inhibition of internalization of NPs after the addition of inhibitors (p < 0.05). b(−) Without affection on NPs after the addition of the inhibitors.

2.7.7. Transcellular Studies. To investigate the transcellular mechanism, transport studies of the NPs were also performed with two pharmacological inhibitors, namely, colchicine (10 μM) and chloroquine (100 μM).26,27 The transport experiments were performed with NPs in the presence of inhibitor for 3 h at 37 °C. Their counterparts in the absence of inhibitor were used as control. Tests were performed in triplicate for each sample. 2.7.8. Paracellular Studies. 2.7.8.1. Immunofluorescence Staining. Immunofluorescence staining was used to investigate the opening of tight junctions mediated by NPs and the influence of C-Jun NH2-terminal kinase (JNK)-specific inhibitor SP (SP600125). Briefly, Caco-2/HT29-MTX cocultured cells were grown to confluence in 35 mm dishes with glass coverslip bottoms, and then treated with tested NPs or NPs with SP, respectively. After incubation with NPs for 3 h, the treated cells were washed three times with prewarmed PBS and then fixed for 20 min at room temperature using freshly paraformaldehyde solution (PFA, 4%). Then the cells were washed thrice with PBS and permeabilized with 0.1% Triton X100/PBS for 15 min at 37 °C. The procedure was repeated and the cells were blocked with 5% normal goat serum in PBS for 60 min at 37 °C. Subsequently, cells were incubated overnight with mouse monoclonal anti-claudin-4 at 4 °C. After washing thrice with PBS, the cells were incubated in rhodamineconjugated goat anti-mouse IgG at 1:100 dilutions in dark. Finally, the cells were rinsed thrice with PBS and visualized 1523

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Figure 1. 1H NMR spectra of TMC (A), TMC−CSK (C), γ-PGA (E), and γ-PGA-Phe (F); FT-IR spectra of TMC (B) and TMC−CSK (D).

using a confocal laser scanning microscope (CLSM, Live 5 DUO, Carl Zeiss, Jena, Germany).28 2.7.8.2. Western Blotting Assays. The cocultured cells were seeded onto 6-well plates and grown to confluence. Then, cells were treated with NPs for 1, 1.5, 2, and 3 h, respectively. After their removal from culture medium, Western blotting for claudin-4 (one of the tight junction proteins) was evaluated at predetermined time intervals (0, 1, 3, 8, and 12 h). Briefly, monolayers of cocultured cells were gently washed twice with ice-cold PBS and then lysed with a lysis buffer. Subsequently, sample was centrifuged at 13,000 rpm at 4 °C, and the supernatant was collected for the subsequent Western blotting analysis.28 Protein concentration was calculated using the Bradford method. Equal amounts of protein samples were separated on 12% SDS−polyacrylamide gels, and the separated proteins were transferred to a nitrocellulose membrane. After blocking in 5% skim milk in Tris-buffered saline containing 0.1% Tween-20 (TBST), the membrane was incubated overnight in blocking buffer with diluted primary antibodies at 4 °C. Subsequently, the membrane was washed three times using TBST, followed

by exposure to alkaline-phosphatase-conjugated secondary antibodies, and then visualized by a BCIP/NBT alkalinephosphatase color development kit (NEL937, PerkinElmer, Boston, MA, USA). Densitometric analysis of specific bands was performed using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). 2.7.8.3. Real-Time PCR Analysis. Caco-2/HT29-MTX cocultured cells were grown to confluence in 6-well plates and then treated with M-NPs or CSK-M-NPs. The gene level expressions induced by those NP treatments and after their removal were investigated. Total RNA was isolated using TRI Reagent (Applied Biosystems, Foster City, CA, USA) following the manufacturer’s instructions and then reverse-transcribed into cDNA using random hexamer and SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Quantitative real-time PCR was performed by the Applied Biosystems 7500 Real-Time PCR System using Power SYBR Green PCR Master Mix (Applied Biosystems) in triplicate for each sample and each gene. The primers used were as follows (5′-3′): Claudin-4 forward, AGAGTGGATGGACGGGTTTAGAGG; Claudin-4 reverse, TGAAGCGGGTGAGGCAGAGAG; glyceraldehyde1524

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Table 4. Characteristics of Tested NPs (n = 3) NPs

size (nm)

PDI

zeta potential (mV)

EE %

DL %

S NPs M-NPs CSK-M-NPs

221.5 ± 4.6 212.2 ± 3.1 198.5 ± 2.1

0.210 ± 0.004 0.245 ± 0.007 0.187 ± 0.009

22.21 ± 0.32 22.31 ± 0.68 19.70 ± 0.51

50.14 ± 5.01 52.78 ± 3.21 57.63 ± 5.02

28.64 ± 2.45 35.26 ± 1.27 38.18 ± 3.04

Figure 2. (A) Release profiles of NPs with different combinations of agents in PBS (pH = 6.8) (mean ± SD, n = 4; *: p < 0.05 vs M-NPs or CSK-MNPs). (B) Percentage of residual amount of insulin in free insulin solution, S NPs, M-NPs, or CSK-M-NPs after certain times of incubation with typsin (mean ± SD, n = 3; #: p < 0.05 vs S NPs or insulin). (C) Cytotoxicity of insulin, M-NPs, or CSK-M-NPs on cocultured cells at the tested concentrations. (D) Cytotoxicity of all inhibitors on cocultured cells at the tested concentrations (mean ± SD, n = 3).

γ-PGA is a natural water-soluble, biodegradable and nontoxic polyamide composed of γ-linked glutamic acid units.29 In this study, γ-PGA was covalently modified with a hydrophobic amino acid, L-phenylalanine (Figure 1E). γ-PGA-Phe was successfully synthesized according to the 1H NMR spectrum (Figure 1F) with Phe grafting degree of 16.01%. Meanwhile, the FITC-labeled insulin was also characterized. 3.2. Preparation and Characterization of NPs. S NPs, cross-linked with TMC and TPP, were inherently unstable in the high ionic strength environment.30−32 Therefore, in the present study, M-NPs were prepared with multiple crosslinking agents, including TPP, MgSO4, and γ-PGA-Phe. TPP and sulfate salts were used to cross-link TMC or TMC−CSK physically by ionic gelation. Meanwhile, physical gelation might also occur between Mg2+ and the carboxylate ions on γ-PGA via an electrostatic interaction.33 Furthermore, the introduction of metal ions can also stabilize the proteins drugs. Besides, L-Phe conjugated to γ-PGA could also enhance the stability of the NPs by reducing the hydration of the NP matrix via hydrophobic interactions. The optimized formulation was composed of 0.64 mg/mL TMC, 0.152 mg/mL insulin, 0.32 mg/mL TPP, 0.26 mg/mL MgSO4, and 0.105 mg/mL γ-PGAPhe (Table 1). The average size of obtained NPs was ∼210 nm with drug entrapment efficiency of 55% and drug loading of 36% (Table 4). The functions of different components added in the NP formulation were demonstrated in a drug release study (Table 2). As shown in Figure 2A, the slowest drug release was observed for the formulation containing all the components,

3-phosphate dehydrogenase (GAPDH) forward, GCACAGTCAAGGCCGAGAAT; GAPDH reverse, GCCTTCTCCATGGTGGTGAA.28 PCR conditions used were denaturation at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 15 and 60 s at 60 °C for annealing and elongation. 2.8. Statistical Analysis. The one-tailed Student’s t-test was performed to compare the two groups by using statistical software (SPSS, Chicago, IL). All data are presented as mean ± SD. A difference of P < 0.05 was considered statistically significant.

3. RESULTS AND DISCUSSION 3.1. Characterization of Synthesized Polymers. NTrimethyl chitosan chloride, a chitosan derivative with trimethylized amine groups, showed superiority over chitosan due to its increasing solubility in physiologic pH conditions. TMC was successful synthesized by substituting the NH2 group into a quaternary ammonium group (Figure 1A,B). According to the results of the 1H NMR spectrum, the degree of quaternization of the synthesized TMC was calculated to be 33.08%. The conjugation of TMC with CSK peptide was confirmed by the peaks at 6.752 and 7.023 ppm for two protons of benzene ring of tyrosine in CSK peptide sequence in the 1H NMR spectrum, and the characteristic peak of benzene ring at 798.15 cm−1 in FT-IR spectrum (Figure 1C,D). According to the results of amino acid detection, the content of CSK peptide in TMC−CSK polymer was calculated to be 0.09 mM/g. 1525

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Figure 3. (A) Cellular uptake of free insulin, M-NPs, or CSK-M-NPs on Caco-2/HT29-MTX cocultured cells or Caco-2 cells, respectively (mean ± SD, n = 3). *: Significant difference from insulin solution on both cell models. #: Significant difference from M-NPs on cocultured cells. (B) Relative cellular uptake percentage (% of control) of insulin, M-NPs, or CSK-M-NPs on Caco-2/HT29-MTX cocultured cells or Caco-2 cells with the addition of free CSK peptide (mean ± SD, n = 3; *: P < 0.05 vs control).

enterocytes, goblet cells, and the mucus layer. The biocompatibility of the prepared NPs and the cytotoxicity of the inhibitors were evaluated. Both M-NPs and CSK-M-NPs showed no significant cytotoxicity at the tested concentrations of insulin (50−400 μg/mL) at an incubation time of 3 h (Figure 2C). No cytotoxicity for inhibitors at tested concentrations was observed for all groups as compared with control group (Figure 2D). 3.5. Internalization Studies of FITC-Insulin from NPs. Our previous study reported the increased interaction of CSK peptide modified single ion cross-linked NPs with HT29-MTX cells and demonstrated the enhanced cellular uptake mediated by the CSK peptide modification.10 In the present study, the cellular uptake of M-NPs and goblet cell-specific peptide modified CSK-M-NPs were investigated on Caco-2/HT29MTX cocultured cells as well as Caco-2 cells. Figure 3A showed the amount of internalized FITC-insulin from tested NPs after incubation of 3 h. Both modified and unmodified NPs exhibited significantly higher uptake compared with free FITC-insulin solution (P < 0.05) in both cell models. The internalization of drugs from CSK-M-NPs (247.10 ± 10.04 μg/mg protein) was 2.21-fold higher than that from M-NPs (112.20 ± 3.05 μg/mg protein) in cocultured cells (P < 0.05), while no significant difference was observed between the two nanoparticles in Caco-2 cells. Moreover, it was found that the presence of free CSK peptide could significantly inhibit the internalization of CSK-M-NPs (Figure 3B, 50%, P < 0.05) in the cocultured Caco-2/HT29-MTX cells, whereas M-NPs were not affected. Besides, free CSK peptide did not influence the internalization

indicating the improved matrix stability of the NPs with multiple agents. 3.3. Enzymatic Stability. To evaluate the protection effect of the nanoparticles on the encapsulated insulin, enzymatic stability of NPs against trypsin was investigated and is shown in Figure 2B. Trypsin is one of the main proteases responsible for insulin degradation in the GI tract. Upon incubation with trypsin at 37 °C, free insulin was rapidly degraded with a halflife (t1/2) less than 60 min, whereas the increased stability of insulin was observed for all NPs. Relatively rapid enzymatic degradation of S NPs occurred with a half-life about 2 h. In comparison, insulin encapsulated in NPs with multiple crosslinking agents was more resistant to enzymatic degradation with prolonged half-life greater than 6 h. After an incubation period of 8 h, almost 80% of the associated insulin was degraded for S NPs, which was 2-fold higher than that for M-NPs or CSK-MNPs. These results demonstrated the greatly enhanced drug protection ability achieved by M-NPs and CSK-M-NPs, which might due to the improved cross-linking strategy. No significant difference was observed between M-NPs and CSK-M-NPs. Therefore, M-NPs and CSK-M-NPs were submitted to further investigation. 3.4. Cytotoxicity Evaluations of NPs and Inhibitors. Goblet cells, the second largest population of epithelial cells, account for 10−24% of the intestinal epithelial cells and are responsible for producing mucus in the GI tract. Goblet celllike, mucus producing HT29-MTX cells were cocultured with Caco-2 cells to simulate the intestinal mucosa containing 1526

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

transferrin indicated that clathrin-mediated endocytosis may be related to the CSK peptide modified NP internalization. 3.7. Specific Inhibition of Endocytosis Pathways of NPs. “Endocytosis” encompasses several distinct mechanisms by which cells internalize macromolecules and particles into transport vesicles derived from the plasma membrane. It controls the entry of particles into the cell and has a crucial role in intercellular communication, signal transduction, and cellular or organismal homeostasis. 34 The importance of the endocytosis pathway is unchallenged. However, its role in ligand modified TMC NPs remains largely unclear. The endocytosis pathway could be inhibited with chemical inhibitors. It needs to be mentioned that one endocytosis pathway consists of different subpathways which may respond to different inhibitors. Therefore, in the present study, a series of chemical inhibitors were used to further investigate the endocytotic pathways as well as the cellular endocytotic subpathways of the tested NPs (Table 3). A mean blocking efficiency of each inhibitor was calculated from the statistical analysis of NPs uptake. Cell membrane is negatively charged due to the presence of heparin sulfate proteoglycans and glycosaminoglycan. Heparinase is an inhibitor of glycosaminoglycan sulfatation and could hydrolyze proteoglycans.35 As shown in Table 4, both M-NPs and CSK-M-NPs were positively charged with zeta potentials of 22 mV. Treatment with heparinases led to a significant decrease of the cellular uptake by 70% and 80% for M-NPs and CSK-MNPs (P < 0.05) (Table 3), respectively, demonstrating that the electrostatic interactions are responsible for an effective cell− NP interaction by improving NP attachment to the negatively charged cell surface. 3.7.1. Macropinocytosis. Macropinocytosis is a type of distinct pathway that nonspecifically takes up a large amount of fluid-phase contents through the mode called fluid-phase endocytosis. Distinct from clathrin-coated vesicles and caveosomes, macropinosomes have no apparent coat structures and are heterogeneous in size.36 Amiloride is commonly used to inhibit macropinocytosis (but not phagocytosis) by specifically blocking the Na+/H+ exchanger and causes strong cytosolic acidification.37 Rottlerin, an inhibitor of protein kinase C, is also responsible for one of the subpathways of macropinocytosis. As shown in Table 3, M-NPs and CSK-MNPs showed a reduction of uptake (P < 0.05) with amiloride or rottlerin, indicating that macropinocytosis was responsible for the internalization of both NPs by the aid of Na+/H+ exchanger and protein kinase C. Interestingly, significant decrease of cellular uptake was only observed for CSK-M-NPs (P < 0.05) after the treatment of cytochalasin D (Cyt D) (Table 3), which could inhibit actin polymerization and membrane ruffling.38 On the contrary, wortmannin, which inhibits phosphatidylinositol-3-kinase (PI3K) during the subpathway of macropinocytosis,39 exhibited a significant inhibition (77%) of M-NPs uptake. This might be due to the subtle conversion of macropinocytosis subpathways after CSK ligand modification. With the CSK peptide modification, the activities of actin polymerization/membrane ruffling might be upregulated, while the PI3K pathway of macropinocytosis was downregulated. 3.7.2. Caveolae-Mediated Endocytosis. Caveolae-mediated endocytosis (CvME) is a type of cholesterol, dynamindependent and receptor-mediated pathway.40,41 The cellular uptake of M-NPs and CSK-M-NPs was significantly decreased (P < 0.05) after the treatment of lovastatin or dynasore (Table

of both NPs in the Caco-2 cell monolayer. These results were consistent with our previous study and suggested that the increased internalization of CSK-M-NPs was mediated by the specific interaction between the CSK peptide and the HT29MTX cells. 3.6. Intracellular Localization of NPs. Due to incomplete understanding of goblet cells and limited reports about CSK peptide, the receptor of CSK peptide on goblet cells is still elusive. The internalization pathway in which CSK peptides increase the cellular uptake of nanoparticles has not yet been reported. Different endocytic markers were used to identify the uptake mechanism of the CSK peptide modified nanoparticles. Transferrin, CTB, and dextran were used as the endocytic markers for clathrin-mediated, caveolae-mediated endocytosis and macropinocytosis, respectively. Specific internalization pathway can be confirmed by the colocalization of nanoparticles and markers in a “pulse−chase” analysis. Fluorescent images of cellular uptake of nanoparticles and endocytic markers are shown in Figure 4. Endocytic markers

Figure 4. Colocalization of NPs with endocytic markers was observed by fluorescence microscopy. Caco-2/HT29-MTX cocultured cell monolayer was incubated with medium containing tested NPs and Alexa Fluor 594-labeled dextran, CTB, or transferrin, respectively. A and B refer to M-NPs and CSK-M-NPs, respectively; red, molecular probes; green, NPs; blue, nucleus.

were conjugated with Alexa Fluor-594 dye (red signal), and nanoparticles were labeled with FITC (green signal). For MNPs, colocalization was observed for dextran and CTB but not for transferrin. However, colocalization with all three markers was observed for CSK peptide modified nanoparticles. These results suggested that different pathways of internalization may exist between the two NPs. Different localization patterns with 1527

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Figure 5. (A) Transport of M-NPs or CSK-M-NPs on Caco-2/HT29-MTX cocultured cells (mean ± SD, n = 3; *, p < 0.05 vs M-NPs; #, p < 0.05 vs CSK-M-NPs). (B) The effect of intracellular trafficking inhibitors on the transport of M-NPs or CSK-M-NPs on Caco-2/HT29-MTX cocultured cells (mean ± SD, n = 3; *, #: p < 0.05 vs controls, M-NPs).

after the CSK peptide modification. The filipin presented subpathway was turned off, while the genistein regulated pathway was turned on. Furthermore, the pathway associated with protein tyrosine kinase might be largely activated or upregulated in the caveolae-mediated endocytosis for CSKmodified TMC NPs. 3.7.3. Clathrin-Mediated Endocytosis. Clathrin-mediated endocytosis (CME) is a kind of receptor-dependent, clathrinmediated, and GTPase dynamin required endocytosis.45,46 The results of intracellular localization (Figure 4) demonstrated that CME may be related to the CSK-mediated internalization. Chlorpromazine is a cationic amphiphilic agent, which can inhibit clathrin-coated pit formation by a reversible translocation of clathrin and its adapter proteins from the plasma membrane to intracellular vesicles.47 Treatment of cells with chlorpromazine induced a significant inhibition (67%) on the internalization for CSK-M-NPs whereas only a slight decrease for M-NPs. Monodansylcadaverine (MDC), another commonly used CME inhibitor, was also applied. A 60% reduction of internalization was observed for CSK-M-NPs (P < 0.05), while no change was observed on M-NPs (P > 0.05). These results were consistent with the observation of intracellular localization study, again implying the involvement of clathrinmediated endocytosis in CSK-M-NPs.

3). These two inhibitors were reported to be involved in cholesterol biosynthesis and reversible inhibition of dynamin GTPase activity. It is worth noting that dynasore is not specific for CvME and may also interfere with other major endocytosis pathways (clathrin- and lipid raft-dependent) by preventing the formation of coat vesicles and the scission of the endocytic vesicles from the plasma membrane.42 Unexpectedly, different results were obtained with filipin and cholesterol oxidase. Filipin is known to inhibit CvME by binding to cholesterol, a major component of glycolipid microdomains and caveolae.43 Cholesterol oxidase was reported to dramatically change the translocation of caveolin to efficiently block CvME. In the present study, filipin treatment resulted in a significant inhibition of M-NPs (P < 0.05) while cholesterol oxidase only inhibited the internalization of CSK-M-NPs (P < 0.05). Another chemical inhibitor, genistein, was used as an inhibitor of protein tyrosine kinase, which catalyzes the tyrosine phosphorylation and activates the CvME signal transduction pathways.44 A moderate inhibition was observed for the internalization of M-NPs (29%), while a much stronger inhibition was gained for CSK-M-NPs (64%). In all, the above-mentioned results demonstrated that CvME was also involved in the uptake of both NPs. However, tremendous changes of their subpathways were demonstrated 1528

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Figure 6. (A) Immunofluorescence stains of the TJs on Caco-2/HT29-MTX cocultured cells after treatment with NPs. (B) Immunofluorescence stains of the TJs on the Caco-2/HT29-MTX cocultured cells after coincubation with NPs and SP. Cell monolayer was stained for claudin-4 (scale bars, 40 μm).

preventing endosomal maturation and subsequently lysosomal degradation of NPs. Chloroquine also provided molecular motors to transport and mature the different endocytic/ phagocytic vesicles.27 As shown in Figure 5B, with the addition of colchicine, the transportation of drugs from M-NPs and CSK-M-NPs was significantly inhibited (P < 0.05), suggesting strong contribution of microtubules for the endocytic transport of both NPs. Meanwhile, M-NPs had a higher inhibition (1.5fold) than CSK-M-NPs (P < 0.05). Chloroquine treatment also significantly affected the cellular transport of M-NPs and CSKM-NPs. Interestingly, compared to the CSK peptide modified NPs, treatment with chloroquine led to a weaker inhibition of drug transport from M-NPs (P < 0.05). All these results demonstrated that transcytosis played an important role in the cellular transport of TMC-based M-NPs and CSK-M-NPs. 3.9. Opening of Tight Junctions and Mechanism. It was generally accepted that chitosan-mediated enhancement of epithelial permeability was caused by the transient disruption of TJs.28 TJs regulated the paracellular pathway, which is a major rate limiting barrier for the drug permeation. The opening of TJs mediated by chitosan was suggested to be related with the transmembrane claudin-4, which is the principal barrierforming protein of TJs and interacts directly with the adaptor proteins. The distributions of claudin-4 on cocultured cell monolayers after the treatment of nanoparticles were investigated with immunofluorescence staining (shown in Figure 6A). A continuous ring appearance of claudin-4 was observed in the control group. After the treatment of M-NPs or CSK-M-NPs, claudin-4 staining at cell−cell contact sites became segmented and discontinuous, and the red fluorescence signal became much weaker, indicating loss of tight junction function. C-Jun NH2-terminal kinase (JNK), a member of the mitogen-activated protein kinase (MAPK) family of signaling protein, was identified as a stress activated protein kinase. JNK activation is an immediate cellular response to osmotic stress. Recent studies indicated that JNK may play an important role in regulation of TJ integrity in different epithelial cells.49 However, there is no report about whether the chitosanmediated openings of TJs were related to the JNK activation or not. Therefore, the immunofluorescence staining of claudin-4

Moreover, the treatment of methyl-beta-cyclodextrin (M-βCD) resulted in mild inhibition for M-NPs (30% reduction) but without affecting CSK-M-NPs, implying absence of the lipid raft-mediated endocytosis after CSK peptide modification. In summary, the results obtained by the colocalization of endocytic markers and specific inhibition study suggested that the modification by CSK peptide could alter the internalization pathways of nanoparticles. The most notable difference was that clathrin-mediated endocytosis was involved in the internalization of CSK-M-NPs but not in unmodified NPs. Besides, although macropinocytosis and caveolae-mediated endocytosis were involved in the internalization of both NPs, the subpathways showed large difference. Some subpathways were “switched on” after CSK peptide modification while the proportion of some others diminished correspondingly. It might be concluded that the endocytosis mechanism of NPs after the CSK peptide modification is complicated and involves multiple pathways. The identification of the receptor of CSK peptide on goblet cells would be favorable for further investigation and application. 3.8. Transport Studies. The transepithelial transport of FITC-insulin loaded nanoparticles was also evaluated on cocultured cell monolayer. The cumulative amount of transported FITC-insulin is presented in Figure 5A. Compared with unmodified NPs, CSK-M-NPs produced a marked increase in the drug transportation from 1 h to 3 h. This improvement could be inhibited by free CSK peptide (p < 0.05). The apparent permeability coefficients of insulin were 2.45 × 10−6 and 3.22 × 10−6 cm·s−1 for M-NPs and CSK-M-NPs, respectively. Two inhibitors which can interfere with the intracellular transport of endosome were applied to investigate their influence on the transport of both NPs. Colchicine inhibits microtubule polymerization by irreversibly binding to tubulin, one of the main constituents of microtubules, and blocking the microtubular-assisted endocytosis. Microtubules form an extensive network throughout the cell cytoplasm and participate in the traffic and transport of various intracellular endocytic vesicles.48 Chloroquine is another inhibitor that accumulates in endosomes/lysosomes, causing the swelling and disruption of endocytic vesicles by osmotic effects, thus 1529

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

Figure 7. (A) Expressions of claudin-4 on Caco-2/HT29-MTX cocultured cells after exposure to tested NPs or subsequent to NP removal for distinct durations. (B) Changes in claudin-4 gene levels on Caco-2/HT29-MTX cocultured cells after exposure to tested NPs or subsequent to NP removal for distinct durations (R refers to after removal of NPs from culture medium, mean ± SD, n = 3; *, P < 0.05 vs control; #, P < 0.05 vs MNPs).

claudin-4 might be one of the reasons for the increased transport of drug from CSK peptide modified NPs. It should be noted that M-NPs showed significantly lower intensity of claudin-4 protein compared with CSK-M-NPs at 3 h (p < 0.05), of which claudin-4 was partially recovered. This phenomenon suggested that CSK ligand modification induced not only a faster opening of TJs but also a quicker recovery than unmodified NPs. In view of safety, disruption of the integrity of TJs in a shorter time interval would be favorable for the oral delivery of hydrophilic macromolecules. The gene transcription of claudin-4 in the cocultured cells after the treatment of nanoparticles was investigated by realtime PCR. The levels of mRNA isolated from cocultured cells were quantified at 1 h, 1.5 h, 2 h, and 3 h after incubation or 1 h, 3 h, and 8 h after NPs’ removal. As shown in Figure 7B, no significant changes of mRNA level were observed for both NPs after 1−3 h of incubation. This result indicated that the reduction of protein level was not mediated by the regulation of mRNA transcription, which was consistent with Sung’s report in which claudin-4 mRNA level remained unchanged in Caco-2 cells after exposure with chitosan.28 However, significant increase of mRNA level was observed for both NPs after their removal and the gene transcription went back to normal

was also performed with the pretreatment of SP, a JNK-specific inhibitor (Figure 6B). Noticeably, the staining of claudin-4 proteins conserved a continuous ring appearance between adjacent cells after the incubation with both tested NPs. This result indicated that the opening of TJs by M-NPs and CSK-MNPs was related to the osmotic stress in a JNK dependent mechanism. 3.10. Expression of Claudin-4 Proteins and Gene Levels. The claudin-4 protein expressions and the claudin-4 mRNA level after the treatment of nanoparticles were investigated quantitatively. At certain time intervals after the incubation and removal of the nanoparticles, claudin-4 expressions in the cocultured cells were investigated by Western blot. As shown in Figure 7A, both nanoparticles induced a reversible reduction of the claudin-4 level which was totally recovered within 8 h after the removal of the nanoparticles. Interestingly, CSK peptide modified NPs resulted in a much quicker reduction of the claudin-4 protein compared to unmodified NPs at 1 h and 1.5 h (p < 0.05) and became comparable with M-NPs at 2 h. This might be ascribed to the increased association of NPs with the epithelial cell surface via goblet cell affinity of CSK peptide and then led to a quickly opening of TJs. The faster reduction of the amount of 1530

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

(7) Chen, M. C.; Mi, F. L.; Liao, Z. X.; Hsiao, C. W.; Sonaje, K.; Chung, M. F.; Hsu, L. W.; Sung, H. W. Recent advances in chitosanbased nanoparticles for oral delivery of macromolecules. Adv. Drug Delivery Rev. 2013, 65 (6), 865−879. (8) Du, W.; Fan, Y.; Zheng, N.; He, B.; Yuan, L.; Zhang, H.; Wang, X.; Wang, J.; Zhang, X.; Zhang, Q. Transferrin receptor specific nanocarriers conjugated with functional 7peptide for oral drug delivery. Biomaterials 2013, 34 (3), 794−806. (9) Yun, Y.; Cho, Y. W.; Park, K. Nanoparticles for oral delivery: Targeted nanoparticles with peptidic ligands for oral protein delivery. Adv. Drug Delivery Rev. 2013, 65 (6), 822−832. (10) Jin, Y.; Song, Y.; Zhu, X.; Zhou, D.; Chen, C.; Zhang, Z.; Huang, Y. Goblet cell-targeting nanoparticles for oral insulin delivery and the influence of mucus on insulin transport. Biomaterials 2012, 33 (5), 1573−1582. (11) Hsu, L. W.; Lee, P. L.; Chen, C. T.; Mi, F. L.; Juang, J. H.; Hwang, S. M.; Ho, Y. C.; Sung, H. W. Elucidating the signaling mechanism of an epithelial tight-junction opening induced by chitosan. Biomaterials 2012, 33 (26), 6254−6263. (12) Hsu, L. W.; Ho, Y. C.; Chuang, E. Y.; Chen, C. T.; Juang, J. H.; Su, F. Y.; Hwang, S. M.; Sung, H. W. Effects of pH on molecular mechanisms of chitosan−integrin interactions and resulting tightjunction disruptions. Biomaterials 2013, 34 (3), 784−793. (13) Sonaje, K.; Chuang, E. Y.; Lin, K. J.; Yen, T. C.; Su, F. Y.; Tseng, M. T.; Sung, H. W. Opening of Epithelial Tight Junctions and Enhancement of Paracellular Permeation by Chitosan: Microscopic, Ultrastructural, and Computed-Tomographic Observations. Mol. Pharmaceutics 2012, 9 (5), 1271−1279. (14) Sieval, A. B.; Thanou, M.; Kotzé, A. F.; Verhoef, J. C.; Brussee, J.; Junginger, H. E. Preparation and NMR characterization of highly substitutedN-trimethyl chitosan chloride. Carbohydr. Polym. 1998, 36 (2−3), 157−165. (15) Akagi, T.; Watanabe, K.; Kim, H.; Akashi, M. Stabilization of Polyion Complex Nanoparticles Composed of Poly(amino acid) Using Hydrophobic Interactions. Langmuir 2009, 26 (4), 2406−2413. (16) Wang, S.; Jiang, T.; Ma, M.; Hu, Y.; Zhang, J. Preparation and evaluation of anti-neuroexcitation peptide (ANEP) loaded N-trimethyl chitosan chloride nanoparticles for brain-targeting. Int. J. Pharm. 2010, 386 (1−2), 249−255. (17) Lin, Y. H.; Sonaje, K.; Lin, K. M.; Juang, J. H.; Mi, F. L.; Yang, H.; Sung, H. W. Multi-ion-crosslinked nanoparticles with pHresponsive characteristics for oral delivery of protein drugs. J. Controlled Release 2008, 132 (2), 141−149. (18) Jin, Y.; Zhou, D.; Yang, H. Y.; Zhu, X.; Wang, X. R.; Zhang, Z. R.; Huang, Y. Effects of degree of quaternization on the preparation and characterization of insulin-loaded trimethyl chitosan polyelectrolyte complexes optimized by central composite design. Pharm. Dev. Technol. 2012, 17 (6), 719−729. (19) Hilgendorf, C.; Spahn-Langguth, H.; Regardh, C. G.; Lipka, E.; Amidon, G. L.; Langguth, P. Caco-2 versus Caco-2/HT29-MTX cocultured cell lines: permeabilities via diffusion, inside- and outsidedirected carrier-mediated transport. J. Pharm. Sci. 2000, 89 (1), 63−75. (20) Yin, L.; Ding, J.; He, C.; Cui, L.; Tang, C.; Yin, C. Drug permeability and mucoadhesion properties of thiolated trimethyl chitosan nanoparticles in oral insulin delivery. Biomaterials 2009, 30 (29), 5691−5700. (21) Harashima, H.; Shinohara, Y.; Kiwada, H. Intracellular control of gene trafficking using liposomes as drug carriers. Eur. J. Pharm. Sci. 2001, 13 (1), 85−89. (22) Hong, S.; Rattan, R.; Majoros, I. J.; Mullen, D. G.; Peters, J. l.; Shi, X.; Bielinska, A. U.; Blanco, L.; Orr, B. G.; Baker, J. R.; Holl, M. M. The role of ganglioside GM1 in cellular internalization mechanisms of poly(amidoamine) dendrimers. Bioconjugate Chem. 2009, 20 (8), 1503−1513. (23) Duchardt, F.; Fotin-Mleczek, M.; Schwarz, H.; Fischer, R.; Brock, R. A comprehensive model for the cellular uptake of cationic cell-penetrating peptides. Traffic 2007, 8 (7), 848−866. (24) Cartiera, M. S.; Johnson, K. M.; Rajendran, V.; Caplan, M. J.; Saltzman, W. M. The uptake and intracellular fate of PLGA

level at 8 h postremoval. This result also suggested that both of the NPs could transiently and reversibly open the epithelial tight junctions.

4. CONCLUSIONS The aim of this research was to elucidate the role of endocytic pathways and opening of tight junctions on the internalization and transport of M-NPs/CSK-M-NPs in Caco-2/HT29-MTX cocultured cells. As the results show, both of the NPs exhibited a superior stability in protecting drugs against the degradation of trypsin compared with S NPs. The internalization mechanism study on cocultured cells with a series of chemical inhibitors and endocytosis markers showed that CvME and macropincytosis were involved with both of the NPs. In addition, the CSK peptide modified NPs were also taken up by cells through CME. Intriguingly, most of the endocytosis subpathways of M-NPs have been altered after CSK peptide modification. Moreover, the immunofluorescence staining, the expression of claudin-4 at protein/gene levels, and the signaling mechanism studies showed that both of the M-NPs and CSKM-NPs could transiently and reversibly open the epithelial TJs by JNK-dependent pathway. CSK-M-NPs could enable a faster opening and also a faster recovery of the TJs compared to MNPs. The higher uptake and transport of the CSK peptide modified NPs in cocultured cells might be attributed to the changes of internalization pathways and the higher activity of opening TJs.

■

AUTHOR INFORMATION

Corresponding Author

*West China School of Pharmacy, Sichuan University, Chengdu 610041, Sichuan, P. R. China. Tel/fax: +86-2885501617. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge financial support from the National Natural Science Foundation of China (81173010).

■

REFERENCES

(1) Sadeghi, A. M. M.; Dorkoosh, F. A.; Avadi, M. R.; Saadat, P.; Rafiee-Tehrani, M.; Junginger, H. E. Preparation, characterization and antibacterial activities of chitosan, N-trimethyl chitosan (TMC) and N-diethylmethyl chitosan (DEMC) nanoparticles loaded with insulin using both the ionotropic gelation and polyelectrolyte complexation methods. Int. J. Pharm. 2008, 355 (1−2), 299−306. (2) Nanish, P.; Ravi, R. P.; Jayvadanm, K. Chitosan mediated targeted drug delivery system: a review. J. Pharm. Pharm. Sci. 2010, 13 (4), 536−557. (3) Mukhopadhyay, P.; Mishra, R.; Rana, D.; Kundu, P. P. Strategies for effective oral insulin delivery with modified chitosan nanoparticles: A review. Prog. Polym. Sci. 2012, 37 (11), 1457−1475. (4) He, C.; Yin, L.; Tang, C.; Yin, C. Size-dependent absorption mechanism of polymeric nanoparticles for oral delivery of protein drugs. Biomaterials 2012, 33 (33), 8569−8578. (5) Dombu, C.; Carpentier, R.; Betbeder, D. Influence of surface charge and inner composition of nanoparticles on intracellular delivery of proteins in airway epithelial cells. Biomaterials 2012, 33 (35), 9117− 9126. (6) Lv, L. Z.; Tong, C.-Q.; Yu, J.; Han, M.; Gao, J. Q. Mechanism of enhanced oral absorption of hydrophilic drug incorporated in hydrophobic nanoparticles. Int. J. Nanomed. 2013, 8, 2709−2717. 1531

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532

Molecular Pharmaceutics

Article

nanoparticles in epithelial cells. Biomaterials 2009, 30 (14), 2790− 2798. (25) Walter, E.; Janich, S.; Roessler, B. J.; Hilfinger, J. M.; Amidon, G. L. HT29-MTX/Caco-2 cocultures as an in vitro model for the intestinal epithelium: in vitro-in vivo correlation with permeability data from rats and humans. J. Pharm. Sci. 1996, 85 (10), 1070−1076. (26) Kruth, H. S.; Jones, N. L.; Huang, W.; Zhao, B.; Ishii, I.; Chang, J.; Combs, C. A.; Malide, D.; Zhang, W. Y.; Zhang, W. Y. Macropinocytosis is the endocytic pathway that mediates macrophage foam cell formation with native low density lipoprotein. J. Biol. Chem. 2005, 280 (3), 2352−2360. (27) Khalil, I. A.; Kogure, K.; Futaki, S.; Harashima, H. High density of octaarginine stimulates macropinocytosis leading to efficient intracellular trafficking for gene expression. J. Biol. Chem. 2006, 281 (6), 3544−3551. (28) Yeh, T. H.; Hsu, L. W.; Tseng, M. T.; Lee, P. L.; Sonjae, K.; Ho, Y. C.; Sung, H. W. Mechanism and consequence of chitosan-mediated reversible epithelial tight junction opening. Biomaterials 2011, 32 (26), 6164−6173. (29) Akagi, T.; Watanabe, K.; Kim, H.; Akashi, M. Stabilization of polyion complex nanoparticles composed of poly(amino acid) using hydrophobic interactions. Langmuir 2010, 26 (4), 2406−2413. (30) Yun, Y.; Cho, Y. W.; Park, K. Nanoparticles for oral delivery: Targeted nanoparticles with peptidic ligands for oral protein delivery. Adv. Drug Delivery Rev. 2012, 65 (6), 822−832. (31) Thumser, A. E.; Storch, J. Liver and intestinal fatty acid-binding proteins obtain fatty acids from phospholipid membranes by different mechanisms. J. Lipid Res. 2000, 41, 647−656. (32) Lopez-Leon, T.; Carvalho, E. L.; Seijo, B.; Ortega-Vinuesa, J. L.; Bastos-Gonzalez, D. Physicochemical characterization of chitosan nanoparticles: electrokinetic and stability behavior. J. Colloid Interface Sci. 2005, 283, 344−351. (33) Nguyen, H. N.; Wey, S. P.; Juang, J. H.; Sonaje, K.; Ho, Y. C.; Chuang, E. Y.; Hsu, C. W.; Yen, T. C.; Lin, K. J.; Sung, H. W. The glucose-lowering potential of exendin-4 orally delivered via a pHsensitive nanoparticle vehicle and effects on subsequent insulin secretion in vivo. Biomaterials 2011, 32 (10), 2673−2682. (34) Conner, S. D.; Schmid, C. S. L. Regulated portals of entry into the cell. Nature 2003, 422, 37−44. (35) Ho, G., Jr.; Broze, G. J.; Schwartz, A. L. Role of heparan sulfate proteoglycans in the uptake and degradation of tissue factor pathway inhibitor-coagulation factor Xa complexes. J. Biol. Chem. 1997, 272, 16838−16844. (36) Sarkar, K.; Kruhlak, M. J.; Erlandsen, S. L.; Shaw, S. Selective inhibition by rottlerin of macropinocytosis in monocyte-derived dendritic cells. Immunology 2005, 116 (4), 513−524. (37) Sasahara, T.; Yayama, K.; Tahara, T.; Onoe, H.; Okamoto, H. Na+/H+exchanger inhibitor augments hyperosmolarity-induced vasoconstriction by enhancing actin polymerization. Vasc. Pharmacol. 2013, 59 (5−6), 120−126. (38) Perumal, O. P.; Inapagolla, R.; Kannan, S.; Kannan, R. M. The effect of surface functionality on cellular trafficking of dendrimers. Biomaterials 2008, 29 (24−25), 3469−3476. (39) Araki, N.; Johnson, M. T.; Swanson, J. A. A role for phosphoinositide 3-kinase in the completion of macropinocytosis and phagocytosis by macrophages. J. Cell Biol. 1996, 135 (5), 1249− 1260. (40) Van der, A. Ma.; Huth, U. S.; Hafele, S. Y.; Schubert, R.; Oosting, R. S.; Mastrobattista, E.; Hennink, W. E.; Peschka-Suss, R.; Koning, G. A.; Crommelin, D. J. Cellular uptake of cationic polymerDNA complexes via caveolae plays a pivotal role in gene transfection in COS-7 cells. Pharm. Res. 2007, 24 (8), 1590−1598. (41) Nichols, B. Caveosomes and endocytosis of lipid rafts. J. Cell Sci. 2003, 116, 4707−4714. (42) Kastl, L.; Sasse, D.; Wulf, V.; Hartmann, R.; Mircheski, J.; Ranke, C.; Carregal-Romero, S.; Martinez-Lopez, J. A.; FernandezChacon, R.; Parak, W. J.; Elsasser, H. P.; Rivera, G. P. Multiple internalization pathways of polyelectrolyte multilayer capsules into mammalian cells. ACS Nano 2013, 7 (8), 6605−6618.

(43) Sadekar, S.; Ghandehari, H. Transepithelial transport and toxicity of PAMAM dendrimers: Implications for oral drug delivery. Adv. Drug Delivery Rev. 2012, 64 (6), 571−588. (44) Nynca, A.; Nynca, J.; Wąsowska, B.; Kolesarova, A.; Kołomycka, A.; Ciereszko, R. E. Effects of the phytoestrogen, genistein, and protein tyrosine kinase inhibitor−dependent mechanisms on steroidogenesis and estrogen receptor expression in porcine granulosa cells of medium follicles. Domest. Anim. Endocrinol. 2013, 44 (1), 10−18. (45) Rappoport, J. Z. Focusing on clathrin-mediated endocytosis. Biochem. J. 2008, 412, 415−423. (46) Pearse, B. M. Clathrin: a unique protein associated with intracellular transfer of membrane by coated vesicles. Proc. Natl. Acad. Sci. U.S.A. 1976, 73 (4), 1255−1259. (47) Wang, L. H.; Rothberg, K. G.; Anderson, R. G. Mis-assembly of clathrin lattices on endosomes reveals a regulatory switch for coated pit formation. J. Cell Biol. 1993, 123, 1107−1117. (48) Elkjaer, M. L.; Birn, H.; Agre, P.; Christensen, E. I.; Nielsen, S. Effects of microtubule disruption on endocytosis, membrane recycling and polarized distribution of Aquaporin-1 and gp330 in proximal tubule cells. Eur. J. Cell Biol. 1995, 67 (1), 57−72. (49) Lin, A. Activation of the JNK signaling pathway: breaking the brake on apoptosis. Bioessays 2003, 25 (1), 17−24.

1532

dx.doi.org/10.1021/mp400685v | Mol. Pharmaceutics 2014, 11, 1520−1532